New paper finds climate models are getting worse rather than better
Via the Hockey Schtick: A paper published today in Geophysical Research Letters finds that the latest climate models are performing even worse than the earlier generations of climate models in predicting
“…both the mean surface air temperature as well as the frequency of extreme monthly mean temperature events due to climate warming.”
The author hypothesizes the reasons for this are that attempts in the latest generation of models to reproduce observed changes in Arctic sea ice are causing “significant and widening discrepancy between the modeled and observed warming rates outside of the Arctic,” i.e. they have improved Arctic simulation at the expense of poorly simulating the rest of the globe. The paper adds to hundreds of other peer-reviewed papers demonstrating the abject failure of climate models.
The paper:
Emerging selection bias in large-scale climate change simulations
Kyle L. Swanson
Abstract:
Climate change simulations are the output of enormously complicated models containing resolved and parameterized physical processes ranging in scale from microns to the size of the Earth itself. Given this complexity, the application of subjective criteria in model development is inevitable. Here we show one danger of the use of such criteria in the construction of these simulations, namely the apparent emergence of a selection bias between generations of these simulations.
Earlier generation ensembles of model simulations are shown to possess sufficient diversity to capture recent observed shifts in both the mean surface air temperature as well as the frequency of extreme monthly mean temperature events due to climate warming. However, current generation ensembles of model simulations are statistically inconsistent with these observed shifts, despite a marked reduction in the spread among ensemble members that by itself suggests convergence towards some common solution.
This convergence indicates the possibility of a selection bias based upon warming rate. It is hypothesized that this bias is driven by the desire to more accurately capture the observed recent acceleration of warming in the Arctic and corresponding decline in Arctic sea ice. However, this convergence is difficult to justify given the significant and widening discrepancy between the modeled and observed warming rates outside of the Arctic.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Perhaps they should try different models for different zones. Each zone has its own rules.
As I noted above, the IPCC general circulation models (GCM) cannot be tuned to correct a fundamental error in the model. The 20th century warming was primarily in the Northern Hemisphere and primarily in high northern latitudes of the Northern hemisphere. This specific warming pattern occurs cyclically in the paleo record and is called a Dansgaard-Oeschger cyclic.
The past D-O cycles were not caused by CO2 changes.
The recent increase in atmospheric CO2 was not predicted to cause warming in the high Northern latitude regions of the earth.
The AGW mechanism predicts that most of the warming should occur in the tropics where there is the most long wave radiation emitted off to space. This has not occurred as tropical region cloud cover increases or decreases, to resist forcing changes, by reflecting more or less short wave solar radiation off into space.
The IPCC general circulation models (GCM) assume the planet amplifies (positive feedback) the AGW forcing change rather than resists (negative feedback) the forcing change. If the planet resists the AGW forcing change (negative feedback) the warming due to a doubling of atmospheric CO2 will be less than 1C.
It is known that the D-O cycles correlate with solar magnetic cycle changes. The D-O warming and cooling cycle is caused by solar magnetic cycle changes. The solar magnetic cycle changes in turn cause a change in the amount of planetary cloud cover low level and high level clouds (cirrus) in higher latitude regions of the planet. There is an interesting series of papers that have been written to explain how the solar cycle changes modulate planetary cloud cover and explain why the effect is different in different regions of the planet. For example, the change in planetary cloud cover is different in the Arctic region than the Antarctic (see my above comment for details.)
The following is a paper that notes there is cyclic variation of North Atlantic temperatures that correlates with solar magnetic cycle length. As the paper notes is a delay in the cooling of 10 to 12 years.
Solar magnetic cycle 24 is the special solar change that causes the cooling phase of a Dansgaard-Oeschger cycle.
As most are aware the magnetic field strength for some unexplained reason is decaying linearly. Due to this change the magnetic ropes that form at the solar tachocline (the solar tachocline is the name for the narrow region in the sun that separates the radiative zone from the convection zone) and then float up to the surface of the sun to form sunspots on the surface of the sun are becoming weaker (less magnetic field strength) and are now starting to be torn to pieces by turbulence as the pass through the solar convection zone. Extrapolation this change the sun is predicted to be spotless in 2017.
The Arctic region cooling is predicted to be the most severe in the winter and the spring. There is now observational evidence that temperatures in the high Arctic have started to cool.
http://arxiv.org/abs/1112.3256
Solar activity and Svalbard temperatures
The long temperature series at Svalbard (Longyearbyen) show large variations, and a positive trend since its start in 1912. During this period solar activity has increased, as indicated by shorter solar cycles. … ….The temperature at Svalbard is negatively correlated with the length of the solar cycle. The strongest negative correlation is found with lags 10 to 12 years.
These models show that 60 per cent of the annual and winter temperature variations are explained by solar activity. For the spring, summer and fall temperatures autocorrelations in the residuals exists, and additional variables may contribute to the variations. These models can be applied as forecasting models. … …We predict an annual mean temperature decrease for Svalbard of 3.5 ±2C from solar cycle 23 to solar cycle 24 (2009 to 2020) and a decrease in the winter temperature of ≈6 C. …. ….A systematic study by Solheim, Stordahl and Humlum [15] (called SSH11 in the following) of the correlation between SCL and temperature lags in 11 years intervals, for 16 data sets (William: solar cycles), revealed that the strongest correlation took place 10 to 12 years after the mid-time of a solar cycle, for most of the locations included. In this study the temperature series from Svalbard (Longyearbyen) was included, and a relation between the previous sunspot cycle length (PSCL) and the temperature in the following cycle was determined. This relation was used to predict that the yearly average temperature, which was -4.2 C in sunspot cycle (SC) 23, was estimated to decrease to -7.8 C in SC24, with a 95% confidence interval of -6.0 to -9.6C [15]. SSH11[15] found that stations in the North Atlantic (Torshavn, Akureyri and Svalbard), had the highest correlations.
William: The following is the Latitude and longitude of Svalbard (Longyearbyen)
78.2167° N, 15.6333° E Svalbard Longyearbyen, Coordinates
As most are aware based on an analysis cosmogenic isotopes the sun in the 20 th century was in the highest solar magnetic activity in 8000 years and the period of highest activity has the longest in 11,000 years. Following this extreme high period of solar magnetic cycle activity there observational evidence that the solar magentic cycle has been interrupted. This specific pattern inn solar magnetic cycle changes is (assuming I understand the mechanisms) what causes a Heinrich event. A Heinrich event is a D-O cycle that is followed by a geomagnetic excursion. (The first phase observed is the normal cooling of a D-O cycle that is then followed by the restart of the solar magnetic cycle. It is the restart of the solar magnetic cycle that caused the geomagnetic excursion which in turn causes the long term severe cooling of the planet, both hemispheres. How the restart of the solar magnetic cycle affects the geomagnetic field is dependent on the orbital parameters at the time of the restart of the solar magnetic cycle.)
http://cc.oulu.fi/~usoskin/personal/nature02995.pdf
Unusual activity of the Sun during recent decades compared to the previous 11,000 years by S. K. Solanki, I. G. Usoskin, B. Kromer, M. Schussler & J. Beer
Here we report a reconstruction of the sunspot number covering the past 11,400 years, based on dendrochronologically dated radiocarbon concentrations. We combine physics-based models for each of the processes connecting the radiocarbon concentration with sunspot number. According to our reconstruction, the level of solar activity during the past 70 years is exceptional, and the previous period of equally high activity occurred more than 8,000 years ago. We find that during the past 11,400 years the Sun spent only of the order of 10% of the time at a similarly high level of magnetic activity and almost all of the earlier high-activity periods were shorter than the present episode. Although the rarity of the current episode of high average sunspot numbers may indicate that the Sun has contributed to the unusual climate change during the twentieth century, we point out that solar variability is unlikely to have been the dominant cause of the strong warming during the past three decades.
William: The authors of the above paper only considered total solar irradiation TSI which is not the major mechanism by which the sun modulates planetary temperature in Northern Atlantic regions. The mechanism is modulation of low level and high level clouds. There are another set of specialists who have almost worked out the details as to how the sun modulates planetary cloud cover. I will when there is unambiguous cooling present a summary of the research and fill in what I believe is the last and key missing mechanism.
Greenland ice sheet temperatures last 11,000 years. This is a graph from Richard Alley’s paper that shows temperatures on the Greenland ice sheet. The D-O cycle is clearly evident. As noted there is no correlation in Greenland ice sheet temperatures and atmospheric CO2 levels in the past. The warming of the Greenland Ice sheet observed in the 20th century was not caused by the increase in atmospheric CO2. The warming was caused by the solar magnetic cycle change and is the same mechanism that caused the past D-O cycles.
http://www.climate4you.com/images/GISP2%20TemperatureSince10700%20BP%20with%20CO2%20from%20EPICA%20DomeC.gif
Perhaps climate scientists should work for a living and get outside to get that much needed data instead of playing on their computers.
The natural decline of a bureaucracy overrun with rent-seekers. They don’t have to provide a service to live high on the hog, so they don’t. This is getting amusing.
Nature 1, BS 0. My wish in life is to see the whole fraud being taken down and the rent seekers and political manipulators getting their just deserts, although most people for some reason do not approve of the punishment fitting the crime. Remember, the climate conspiracy is beyond high treason, The official punishment for plain boring old high treason is for men, to be hung ,drawn and quartered, for women, to be burned at the stake.
Greg Goodman says:
May 16, 2013 at 5:50 pm
Here, this is one thing I bet they don’t get out of the models:
Rate of change of atmospheric CO2 in the middle of the Pacific matches AO index.!
http://climategrog.wordpress.com/?attachment_id=231
[…]
Now does that mean AO is driving global CO2 variation or is this a case of common cause?
——————————————————————————————————————-
Or does it mean that third category of explanation that people (especially climate scientsists) seem to forget – sheer blind coincidence?
Looking at you link, my eyes (like everyone elses, hard-wired to find patterns) say “too good for a fluke” but it’s my brain’s job to over-rule them and allow that possibility (however slight) until its ruled out.
cotwome quotes a weather forecast for “snowfall NEVER BEFORE SEEN” this late in the season.
Surely, this must be some new usage of the word “NEVER” (even in all caps!) Perhaps it would only take a brief perusal of anecdotal weather reports from 100 years ago or thereabouts. Or perhaps the writer has never seen it before. Maybe it means never before seen since Alaska has been a State (because it didn’t matter before then?)
Either way, the whole “never before seen, ever in the history or prehistory of humanity” meme is getting a bit old. Maybe they could spice it up with a popular movie image or a picture of a cute cat.
shepherdfj says:
May 16, 2013 at 4:30 pm
“They have an out you know. All they have to admit to is that after a certain point, CO2′s effect as a greenhouse gas diminishes greatly, and that point is now reached; and this is why the warming stopped. But then, if they admitted this, then they would be denying their basic premise. Hmmmm, there are other greenhouse gases to pick on, surely. How about methane or nitrous oxide… do I hear a bid for ozone?”
If they could change us all over to worrying about our Nitrous Oxide footprint, they would be “laughing” all the way to the bank!
This is not the first time that modelling has encountered “An Inconvenient Truth”.
Tim Flannery, Robert Manne and The Australian Broadcasting Commission really should wisen up to themselves.
The important point to take home is that because there are more climate models today than there were 10 years ago, and they all say the temperature is rising, then there is a consensus which is becoming stronger. Any errors or problems in the models will be averaged out by sheer weight of numbers and we can trust the consensus result.
If I can get enough people to tweet 1+1 = 3, then eventually we will have a consensus and we can lobby government for funding to encourage action to account for this Hyper Addition Phenomena, which is caused at least in part by man-made-global warming.
What the models builders ignore is that they, the model builders are one of the inputs to the model. The models are not predicting the future – that is very difficult to do. What the models are actually predicting is what the model builders believe the future will be – that is a much simpler problem.
The model builders believe the models are correct because the models are telling them what they already believe to be true. If the models were actually able to predict the future, the model builders would not believe the results, and would “fix” the models.
Think about it. You are a model builder and believe that increasing CO2 will lead to increasing temps. You run your model and it shows otherwise – it predicts declining temps. You will immediately think your model has a problem and will adjust it. However, what if in reality the model was correct, that temps will decline? You have just broken your model for the “mistake” of successfully predicting the future.
Thus, few if any models have any hope of predicting the future, because unless the future coincides with what the model builder believe the future to be, the model build will “fix” the model so that it coincides with his/her belief. Thus, building model to predict the future is near hopeless, because the model builder’s themselves get in the way of the validation process.
John Law says:May 17, 2013 at 4:56 am
[…]
If they could change us all over to worrying about our Nitrous Oxide footprint, they would be “laughing” all the way to the bank!
Imminent Banning of Viagra and Cialis. Helps with population control too! /sarc
cotwome says:
May 16, 2013 at 7:34 pm
Meanwhile… south of the Arctic Circle:
URGENT – WINTER WEATHER MESSAGE…
Since we are almost 2 months into spring when do we get rid of the “Winter” in storm warnings. Heck we are 2 weeks from meteorological summer.
Does this get a name?
And in news just in:
Hansen is partly right. The ‘warming has not stalled’. In that he is correct because the warming is over, finished, kaput.
ferd
‘The model builders believe the models are correct because the models are telling them what they already believe to be true. ‘
Wrong. You’ve never worked as a model builder. Or if you have, you didnt know what you were doing.
@ur momisugly Steven Mosher says: May 17, 2013 at 11:36 am
‘The scientists believe the cAGW hypothesis is correct because the models are telling them what they already believe to be true. ‘
I think ferd would approve.
Computer models are simply implementations of numerical methods to solve mathematical equations, which are themselves reprentations of physical processes that scientists believe to be true. Some are done poorly e.g. NASA/GISS Model E…
@ur momisugly William Astley….several days In this post, http://wattsupwiththat.com/2013/05/10/why-reanalysis-data-isnt-2/#comment-1303136, I made a comment in response to another comment regarding a graph that shows sst changes from 1955 to present, along with 3 volcanic events. lgl had said no one can look at Figure 4 and see any effect on sst from the eruptions. So…..”If you look approximately 6 years out past each eruption, the graph shows a large upward heat spike on the surface:ICOADS SST line. That spike following all three eruptions gains approximately 2.5C from the point where the ICOADS line crosses the eruption event to the peak of the ICOADS line 6 years out. Is this just a coincidence?” -That is my first thought followed by this after Greg Goodman stated that he had mentioned this 2 years earlier……..
goldminor says:
May 12, 2013 at 2:53 pm
@ur momisugly Greg Goodman…then the likely interaction should have something to do with a diminished cloud cover over the ocean at 6 years out? Could the fallout of the last particles draw other particles out of a region of the atmosphere and so lead to a clearer than normal atmosphere for a period of time?……..What I didn’t say and should have added is ‘What if the fallout of the volcanic debris changes the charge of the atmosphere?’. A change that could make the atmosphere transparent or more transparent than it would normally be to some wavelength of incoming.
Then this was my final thought…..”Couldn’t this be the smoking gun for why temperatures rose so high above average in the first place? Global warming has been volcano-induced. For what ever natural forces were in play, with a {slight boost?} from CO2, was then amplified by this volcanic aftereffect. So you have a ‘hot’ sun and a Windex clear atmosphere for it to penetrate and cause { add to} the great El Nino of 1997/98. All of the heat from that event has been dissipating ever since. Isn’t that why the extra warmth in the northern Atlantic encompassed so much of Greenland as well as warming Europe for the 10 years after 1998?”. The Pinatubo event, in particular, strikes with perfect timing to where the 6 year suspected atmospheric effect coincides with the solar rise after the {bottom of the} minimum. In the first 2 earlier volcanic events the 6 year effect coincided with the top of the solar max. Wishful thinking overall on my part. Although, it sure looks like a potential suspect for the warming.
Justthinkin says:
I really get tired of calling these guys “scientists” let alone “climate” ones.
Personally I like “climatologist”. Kinda reminiscent of “Scientologist”
There is a divergence problem here, too, as we know, but not one they can apply “Mike’s Nature trick” to. Really, the two divergence problems smell of the same problem: assumptions. This paper seems to address assumptions and the collective “We Want Warming!” meme – a herd mentality applied to the assumptions.
By this time, some of the modelers should have some iterations telling them their assumptions are not holding true anymore and should have at some point built in some weighted, automatically-popping-in (currently) lesser assumptions. The question has to be, “Why not?” Do they individually not want to be the first to break the assumption logjam? Is that why the now older and out-of-date assumptions keep on being used?
The current non-rise/flattening out is as long or longer than the previous rise upon which the models were being written during the ’80s and ’90s, so there really isn’t a lot of excuse for not revising the code and/or the assumptions.
If skeptics were stupid enough to linearly regress the post-1998 flattening and apply a straight-line regression, the skeptics could project a year 2100 temperature. But they aren’t that dumb: they know it will up-turn at some point and down-turn at others, so a straight linear projection out that far would be, well, just silly.
Steve Garcia
@TonyG – That is actually pretty funny.
Suggestion to Anthony: Should skeptics make a Climategate ‘based on real life’ movie? If so, should Tom Cruise play Michael Mann? Subtitled “Mission Improbable” perhaps…
Steve Garcia
@frank K. May 17, 2013 at 12:24 pm:
To be real scientific models the models need to be based on “physical processes that the scientists” KNOW “to be true.” I’ve said here before that when that is the case, they are real models (i.e., NOT GIGO). When the model is done re an area NOT known for sure to be true, when the inputs are merely assumptions, there has to be a distinction made about the model.
All models are not necessarily representations of real physical processes. It is up to the modeler(s) to inform the audience (in ALL output graphs and tables) of any processes or “constants” which have been assumed.
Steve Garcia
feet2thefire says:
May 17, 2013 at 1:36 pm
Models are (and will always be) mere representations of reality – sometimes crude representations. And, unfortunately, most climate models are mixtures of all different kinds of physical models – atmospheric dynamics models, ocean circulation models, sea ice models, radiation models, aerosol and atmospheric chemistry models, etc. So they end up being HUGE computer codes which are both highly coupled and highly NON-LINEAR. Non-linear means that there is NO guarantee that a meaningful numerical solution is possible for a given set of initial/boundary conditions.
Yet, we see press release after press release where modelers are using their “projections” to scare people about climate change…and that is flat-out WRONG.
Steven Mosher says:
May 16, 2013 at 9:31 pm
“Wrong. people always forget satillites and bouys and think that GISS is the only record.”
The problem with all those nice buoy records is that those buoys tend to be fairly peripatetic, that is they don’t stay in one place for long. It’s hard to do a temperature database when your thermometers are doing a daily site relocation. For instance, if you go to NOAA’s North pole webcam site
http://www.arctic.noaa.gov/gallery_np.html
Then scroll down to the line ” weather records” and click thru the annual records you’ll find drift maps for the installations, which were usually placed within a degree of Lat 90 N, but the majority ended there duty somewhere between Lat 65 to Lat 70 off the East coast of Greenland.